Universal HIV testing and the impact of late diagnosis on disease stage among adults in urban Ethiopia

Abstract

Abstract Background Treatment as prevention evolved into the universal HIV test-and-treat (UTT) strategy, which entails testing to the general population and treatment to every people living with HIV. We investigated universal testing (UT) performance and its determinants in urban Ethiopia and explore magnitude of late diagnosis and its impact on disease stages. Method We used data from the Ethiopia Population Based HIV Impact assessment (EPHIA), conducted in 2017/2018 which was a cross-sectional and household-based study. For current analysis, we considered self-report first diagnosis to estimate universal testing irrespective of their serostatus and also consider HIV LAg avidity vs viral load vs plasma antiretroviral drug level algorithm to categorize the late diagnosis. We finally evaluate disease stages using CD4 count and viral load. A 2-level multilevel mixed-effect logistic regression model was employed. The effects of individual-level predictors were quantified by the estimates from the fixed-effect part of the model with p-value < 0.05. Result Data were collected from 18,926 adults among those 29.4% of people living in Urban Ethiopia were never tested for HIV. Never tested females was 26.4% (95% CI = 25.3; 27.5). Never tested among divorced and widowed were 19.4% (95% CI: 17.3; 21.8) and 28.3% (95% CI: 24.6; 32.2), respectively. Never tested among elderly and youth were high (28.3% among 45–54 years old) to (41.2% among 55–64 years old) to 47.8% among 15–24 years old. Overall, late HIV diagnosis among adults in urban Ethiopia was 25.9% (95% CI: 21.7, 30.2). Late diagnosis varies by region ranged from 38.1% in the Gambella to 5.8% in Benishangul Gumuz. Advanced immune suppression (CD4 count < 350 cells/µl) among newly diagnosed long-term infection were significantly higher compared to those who were recently infected which accounted 47.8% (95%CI = 33.2–52.1) and 30.9% (95%CI = 21.3–32.2), respectively. Moreover, Viral load suppression were significantly lower among those who were late diagnosed 26.1% (95%CI = 13.6–33.8) compared to those of newly infected 89.6% (95%CI = 76.2; 93.4). Conclusion With the aim of UT for high risk and priority population, the low rate of HIV testing among widowed, elderly, young adolescent and women in urban Ethiopia calls for enhanced HIV testing. Moreover, the low HIV testing and high late diagnosis among the high-burden regions calls for region-specific intervention. Advanced disease stages as a result of the high proportion of late diagnosis may impact on fueling community transmission and hinder treatment outcome among PLHIV.